Four questions are addressed in this proposal: (1) What percent of extra-lung lymph contaminates the chronic lung lymph fistula model? (2) Does the bronchial circulation contribute to lung lymph? (3) In models of lung injury, do the relative proportions of bronchial and pulmonary contributions to lung lymph change? (4) Is there a mixed venous chemoreceptor in the lung and does it participate in the control of breathing? Each of these questions can be answered by the use of the goat model in association with various forms of extracorporeal perfusion (systemic and/or pulmonary bypass). The first three questions address in a direct fashion the composition of lung lymph in health and disease. Information on the contributions to lung lymph will assist in the interpretation of animal studies of lung injury. Similarly, the determination of the vascular site of injury in specific disease processes will influence therapy directed at that injury. The fourth question will help to resolve the mechanism which drives respiration, and the poorly understood process which couples metabolic work to ventilatory output.